Field of the Invention
The present invention relates to a thin film actuated mirror array in an optical projection system and to a method for manufacturing the same, and more particularly to a thin film actuated mirror array in an optical projection system for preventing a mis-operation of an actuator due to a photo leakage current caused by an incident light, and to a method for manufacturing the same.
In general, light modulators are divided into two groups according to their optics. One type is a direct light modulator such as a cathode ray tube (CRT) and the other type is a transmissive light modulator such as a liquid crystal display (LCD). The CRT produces superior quality pictures on a screen, but the weight, the volume and the manufacturing cost of the CRT increase according to the magnification of the screen. The LCD has a simple optical structure, so the weight and the volume of the LCD are less than those of the CRT. However, the LCD has a poor light efficiency of under 1 to 2%, because the incident light has polarization. Also, there are some problems in the liquid crystal materials of the LCD such as sluggish response and overheating.
Thus, a digital mirror device (DMD) and actuated mirror arrays (AMA) have been developed to solve these problems. Now, the DMD has a light efficiency of about 5% while the AMA has a light efficiency of above 10%. The AMA enhances the contrast of a picture projected on a screen, so the picture is more apparent and brighter. The AMA is not affected by the polarization of rays of light incident. Also, the AMA does not affect the polarization of a reflected light as well. Therefore, the AMA is more efficient than the LCD or the DMD.
FIG. 1 shows a schematic diagram of an engine system of a conventional AMA disclosed in U.S. Pat. No. 5,126,836 (issued to Gregory Um). Referring to FIG. 1, light incident from a light source 1 passes through a first slit 3 and a first lens 5 and is divided into red, green, and blue lights according to the Red.Green.Blue (R.G.B) system of color representation. After the divided red, green, and blue lights are respectively reflected by a first mirror 7, a second mirror 9, and a third mirror 11, the reflected lights are respectively incident on AMA devices 13, 15 and 17 corresponding to the mirrors 7, 9 and 11. The AMA devices 13, 15 and 17 respectively tilt the mirrors installed therein, so the rays of light incident are reflected by the mirrors. In this case, the mirrors installed in the AMA devices 13, 15 and 17 are tilted according to the deformation of active layers formed under the mirrors. The rays of reflected light by the AMA devices 13, 15 and 17 pass a second lens 19 and a second slit 21 and form a picture on a screen (not shown) by means of a projection lens 23.
In most cases, ZnO is used as a material forming the active layer. However, lead zirconate titanate (PZT:Pb(Zr,Ti)O.sub.3) has been found to have a better piezoelectric property than ZnO. PZT is a complete solid solution made of lead zirconate (PbZrO.sub.3) and lead titanate (PbTiO.sub.3). At a high temperature, PZT exists in a paraelectric phase whose crystal structure is a cubic. While at a room temperature, PZT exists in an antiferroelectric phase whose crystal structure is an orthorhombic and in a ferroelectric phase whose crystal structures are a rhombohedral or a tetragonal according to the composition ratio of Zr and Ti.
PZT has a morphotropic phase boundary (MPB) of the tetragonal phase and the rhombohedral phase where the composition ratio of Zr and Ti is 1:1. PZT has a maximum dielectric property and piezoelectric property at the MPB. The MPB does not lie in a specific composition ratio, but lies over a relatively wide region where the tetragonal phase and the rhombohedral phase coexist. The phase coexistent region of PZT is reported differently depending on researchers. Various theories such as thermodynamic stability, compositional fluctuation, and internal stress have been suggested as the reason for the phase coexistent region. Nowadays, a PZT thin film can be manufactured by various processes such as a spin coating method, a chemical vapor deposition (CVD) method, or a sputtering method.
The AMA is generally divided into a bulk type AMA and a thin film type AMA. The bulk type AMA is disclosed in U.S. Pat. No. 5,469,302 (issued to Dae-Young Lim). The bulk type AMA is formed as follows. A ceramic wafer having a multilayer ceramic where metal electrodes are inserted is mounted on an active matrix having transistors. After sawing the ceramic wafer, a mirror is mounted on the ceramic wafer. However, the bulk type AMA has some disadvantages that very accurate process and design are required, and the response of an active layer is slow. Therefore, the thin film AMA that can be manufactured by using semiconductor manufacturing technology has been developed.
The thin film AMA is disclosed at U.S. Ser. No. 08/602,928 entitled "THIN FILM ACTUATED MIRROR ARRAY FOR USE IN AN OPTICAL PROJECTION SYSTEM", which is now pending in USPTO and is subject to an obligation to the assignee of this application.
FIG. 2 shows a cross-sectional view of the thin film AMA. Referring to FIG. 2, the thin film AMA has an active matrix 60 and an actuator 90 formed on the active matrix 60. The active matrix 60 has a substrate 50 having M.times.N (M, N are integers) numbers of transistors (not shown) therein, M.times.N (M, N are integers) numbers of connecting terminals 53 respectively formed on the transistors, a passivation layer 56 formed on the substrate 50 and on the connecting terminal 53, and an etch stop layer 59 formed on the passivation layer 56.
The actuator 90 has a supporting layer 68, a first electrode 71, an active layer 74, a second electrode 77, and a via contact 80. The supporting layer 68 has a first portion attached to the etch stop layer 59 under which the connecting terminal 53 is formed. Also, the etch stop layer 59 has a second portion parallely formed above the bottom of the active matrix 60. The first portion of the supporting layer is called an anchor 68a. An air gap 65 is interposed between the second portion of the supporting layer 68 and the etch stop layer 59. The first electrode 71 is formed on the supporting layer 68, the active layer 74 is formed on the first electrode 71, and the second electrode 77 is formed on the active layer 74. The via contact 80 is formed from a portion of the active layer 74 under which the connecting terminal 53 is formed to the connecting terminal 53. The via contact 80 connects the first electrode 71 with the connecting terminal 53.
A manufacturing method of the thin film AMA will be described below.
FIGS. 3A to 3D illustrate the manufacturing steps of the thin film AMA shown in FIG. 2. In FIGS. 3A to 3D, the same reference numerals are used for the same elements in FIG. 2.
Referring to FIG. 3A, at first, a substrate 50 in which M.times.N number of transistors (not shown) are mounted and M.times.N number of connecting terminals 53 respectively formed on the transistors are provided. Subsequently, a passivation layer 56 is formed on the connecting terminal 53 and the substrate 50. The passivation layer 56 is formed by using a phosphor-silicate glass (PSG) and by a chemical vapor deposition (CVD) method so that the passivation layer 56 has a thickness of between 0.1 .mu.m and 2.0 .mu.m. The passivation layer 56 protects the substrate 50 having the transistors during subsequent manufacturing steps.
An active matrix 60 is completed after an etch stop layer 59 is formed on the passivation layer 59. The active matrix 60 includes the substrate 50, the connecting terminal 53, the passivation layer 56, and the etch stop layer 59. The etch stop layer 59 is formed by using nitride and by a chemical vapor deposition method so that the etch stop layer 59 has a thickness of between 1000 .ANG. and 2000 .ANG.. The etch stop layer 59 prevents the passivation layer 56 and the substrate 50 from etching during subsequent etching steps.
A sacrificial layer 62 is formed on the etch stop layer 59. The sacrificial layer 62 is formed by using a phosphor-silicate glass and by a chemical vapor deposition method so that the sacrificial layer 62 has a thickness of between 1.0 .mu.m and 2.0 .mu.m. In this case, the flatness of the surface of the sacrificial layer 62 is poor because the sacrificial layer 62 covers over the substrate 50 having the transistors. Thus, the surface of the sacrificial layer 62 is planarized by using a spin on glass (SOG) or by a chemical mechanical polishing. Subsequently, the sacrificial layer 55 is patterned to expose a portion of the etch stop layer 59 where the connecting terminal 53 is formed thereunder. An anchor 68a will be formed at the exposed portion of the etch stop layer 59.
Referring to FIG. 3B, a first layer 67 is formed on the exposed portion of the etch stop layer 59 and on the sacrificial layer 62. The first layer 67 is formed by using nitride. The first layer 67 is formed by a sputtering method or a CVD method so that the first layer 67 has a thickness of between 0.1 .mu.m and 2.0 .mu.cm. A first electrode layer 70 is formed on the first layer 67 by using a metal such as platinum or tantalum and by a sputtering method or a CVD method so that the first electrode layer 70 has a thickness of between 0.1 .mu.m and 1.0 .mu.m. Next, the first electrode layer 70 is iso-cut so as to separately apply a first signal (picture current signal) to each of pixel including the first electrode layer 70.
A second layer 73 is formed on the first electrode layer 70 by using a piezoelectric material such as lead zirconate titanate (PZT) or an electrostrictive material such as lead magnesium niobate (PMN). The second layer 73 is formed by a sol-gel method, a sputtering method, or a CVD method so that the second layer 73 has a thickness of between 0.1 .mu.m and 1.0 .mu.m. A second electrode layer 76 is formed on the second layer 73 by using a metal such as aluminum or silver and by a sputtering method or a CVD method so that the second electrode layer 73 has a thickness of between 0.1 .mu.m and 1.0 .mu.m.
Referring to FIG. 3C, the second electrode layer 76, the second layer 73, and the first electrode layer 70 are respectively patterned to form a second electrode 77, an active layer 74, and a first electrode 71. Thus, M.times.N number of pixels having predetermined shapes are formed. At that time, a portion of the active layer 74 is exposed by etching a portion of the second electrode 77 where the connecting terminal 53 is formed thereunder. Portions of the active layer 74, the first electrode 71, the first layer 67, the etch stop layer 59, and the passivation layer 56 are etched. Subsequently, a via hole 79 is formed from the exposed portion of the active layer 74 to the connecting terminal 53.
Referring to FIG. 3D, a via contact 80 is formed in the via hole 79 by filling the via hole 79 with an electrically conductive material, for example, tungsten. The via contact 80 is formed by a sputtering method or a CVD method. The via contact 80 connects the connecting terminal 53 to the first electrode 71. The first signal transmitted from outside is applied to the first electrode 71 through the transistor, the connecting terminal 53, and the via contact 80. While, a second signal (bias current signal) transmitted from outside is applied to the second electrode 77 through a common line (not shown). Therefore, an electric field is generated between the second electrode 77 and the first electrode 71. The active layer 74 formed between the second electrode 77 and the first electrode 71 is deformed by the electric field. The active layer 74 is deformed in perpendicular direction to the electric field, so the actuator 90 including the active layer 74 is actuated upward by a predetermined angle. The second electrode 77 is also tilted upward, and the second electrode 77 reflects the incident light from the light source (not shown) by a predetermined angle.
Subsequently, the first layer 69 is patterned to form a supporting layer 68 for supporting the actuator 90. A portion of the supporting layer 68 is attached to the etch stop layer 59 where the connecting terminal 53 is formed thereunder. The attached portion of the supporting layer 68 is called an anchor 68a. After the sacrificial layer 62 is removed by using a hydrogen fluoride vapor, pixels are rinsed and dried to complete the thin film AMA.
However, in the above-described thin film AMA, the light incident from a light source is projected to the second electrode 77 as well as the other area except the second electrode 77. Thus, a photo leakage current caused by the incident light flows through the active matrix 60. Due to the photo leakage current, the actuator 90 is mis-operated before the first signal is applied or during the actuator 90 is actuated.